The present application addresses the general problem of antibiotic resistance, in particular antibiotic-resistant tuberculosis (TB). The long-term goal of this research program is to develop new DNA gyrase inhibitors that are highly effective anti-tuberculosis (TB) agents with susceptible, multidrug-resistant, extensively drug- resistant, and completely drug-resistant forms of the disease. The specific goal of the present application is to develop a novel anti-mutant strategy for identifying lead antibacterial agents that will severely restrict the emergence of resistant mutants and to exploit the strategy to specifically identify anti-mutant gyrase inhibitors that are active against wild-type and quinolone-resistant mutants of M. tuberculosis. The anti-mutant strategy is based on the central hypothesis that new antibiotics having a closed or very narrow mutant selection window will severely restrict the rate at which resistant mutants emerge. Such antibiotics are expected to have increased clinical utility and longevity of use. Preliminary studies show that anti-mutant activity of new gyrase inhibitors does narrow the mutant selection window and restricts the recovery of induced mutants. Three specific aims have been developed to integrate medicinal chemistry, microbiology, biochemistry and molecular modeling studies into a new anti-mutant strategy and to identify lead structures for development as effective anti-TB agents. In Aims 1 and 2 we will develop new dione-class and C-7 aryl-quinolone-class agents that are active against wild-type cells, known quinolone-resistant gyrase mutants, and induced mutants generated against progenitor compounds within each class. In a third aim we will exploit two different models for agents binding to gyrase to identify and characterize new types of anti-mutant gyrase inhibitors. The expected outcome of this work is the generation of new lead gyrase inhibitors that exhibit a narrow mutant selection window and severely restrict the recovery of resistant mutants. Successful completion of these studies will provide a general framework for obtaining agents with excellent anti-mutant activity. The framework is expected to apply to many other structural and mechanistic antibiotic classes.